Settle, stay, wait, leave it and drop it
A mat settle, stay and wait built one D at a time, and leave it and drop it with food and toys, with a clear stop where guarding begins.
- Start here: the essentials
- Understand: settle on a mat
- Duration, distraction and distance
- Stay, wait and the release word
- Leave it and drop it
- Doors, bowls and impulse control
- The relaxation protocol, benchmarks and the 80-percent rule
- See it: Pearl and the condo kitchen
- Practice
- Your dog
- Check your work
- Knowledge check and scenario
Start here Essentials
At minimum, teach the dog to go to a mat and lie there for ten seconds, start leave it with a treat under your hand, and write down the longest settle.
- Mark any step onto the mat. Put the mat down and mark and treat each time a paw lands on it, about ten times (settle on a mat).
- Lure a down on the mat and feed there. Deliver every treat on the mat, between the front paws (settle on a mat).
- Build three, five, then ten seconds. Stand beside the mat, feed during the down, then say the release word (the release word).
- Start leave it with a covered treat. If the dog shows none of the guarding signs, cover a treat with your hand and mark the moment the dog looks away (leave it).
- Record the longest settle. Write the seconds in the sessions sheet and the stage in the cue inventory (the record).
Minimum time: about four sessions of 2 to 5 minutes over a few days. Done when: the dog lies on the mat for ten seconds with you standing beside it, and the cue inventory says so.
Settle on a mat Understand
A mat settle is a dog going to a particular mat or bed and lying there, relaxed, until released. It is the most useful behavior in this chapter for daily life. A dog on its mat is not jumping on guests, begging at the table, getting under feet while you cook or pacing while you work. It is also a portable place: once the dog has learned the mat in the living room, the same mat can go to a friend's house, a café patio or the veterinarian's waiting room.
Three things make the mat work. The dog learns that going to the mat pays. It learns that staying on the mat pays, and keeps paying while it stays. And it learns that the release word, not your walking away or the doorbell, ends the job. Each of those is taught separately, and in that order.
Getting the dog onto the mat
Two published methods start differently and end in the same place.
Karen Pryor Clicker Training's mat page, written by a certified trainer, uses shaping: rewarding small steps toward the final behavior without luring. The handler stands still behind the mat and says nothing. The moment a paw lands on the mat by chance, the handler clicks and tosses a treat off the mat, so the dog has to leave and come back for another try. Then the handler waits for two paws, then three, then all four. Once four paws land reliably, the handler waits for the dog to sit on the mat without a cue, and then for it to lie down. For a dog new to the game, the page suggests clicking and treating ten times in a row before raising the bar. Preference
The AKC's "go to your place" page, also written by a certified trainer, starts by clicking any interest in the mat (a sniff, a step) and shaping to four paws. Then it adds the release word, lures the dog off the mat with a tossed treat, and waits to see if the dog goes back. Next, it lures the dog into a down on the mat before rewarding. Duration, distance and distractions come next, and the cue word ("mat", "place") is added last, once the dog does the whole behavior in any circumstance. Preference
A 2025 client handout from Today's Veterinary Practice, a veterinary journal, gives a third version with numbers. It puts the mat a few feet from the handler, rewards each approach, then tosses treats away from the mat so the dog has to return. When the dog is standing on the mat, the handler asks for a sit or down. The handout adds a word ("mat") only after the dog sits or lies down on the mat on its own 8 times out of 10. Convention
Either way works. Shaping takes a little longer at the start and produces a dog that offers the behavior on its own. Luring the down is faster for a dog that already knows "down" from First behaviors. For a first pass, the Start here route uses the quickest combination: mark any step onto the mat, then lure the down.
Where the reward lands
The methods above toss treats off the mat at the start. That is a reset: it gets the dog off the mat so it can practice getting back on. Once you are building the down and the stay, change where the treat lands. Deliver it on the mat, between the dog's front paws, while the dog is lying down. The reward for staying arrives at the mat, not away from it. A dog that is fed off the mat while you are building duration learns that the good things happen somewhere else, and gets up to find them. Physical principle
That is the same timing principle as in How dogs learn: the reward reinforces whatever the dog is doing when it arrives. Feed a dog that is lying on the mat, and lying on the mat is what you are paying for.
From lying down to settling
Lying on a mat and being relaxed on a mat are different. A dog can lie in a stiff, ready-to-spring down with its eyes on you and the treat pouch. A settled dog shifts its weight onto one hip, stretches a leg out, lowers its head, rests its chin on its paws or sighs. The Today's Veterinary Practice handout lists these signs (hips shifted to one side, a leg stretched out, a shoulder down, lying on one side, head down, chin between the paws, sighing, eyes closing, slower breathing) and suggests rewarding them as they happen. Convention
Watching for those signs changes how you reward. Early on, you reward the down. Later, you wait for a small relaxation (a hip rolling over, a long breath) and reward that. The dog learns that relaxing on the mat is what pays. Read your dog covers the opposite signals (lip licks, yawns, a tense body); if you see those on the mat, the session is too long or the room too busy.
How long, and how often
Short sessions work better than long ones. The Today's Veterinary Practice handout suggests sessions of 2 to 5 minutes, 2 to 3 times a day, and notes that progress is possible with one 30-second session a day. VCA's page on predictability and enrichment, written by veterinary behaviorists, suggests starting rest training on a mat close to you with a 5- to 10-minute session, and gradually increasing the interval between treats while the dog stays resting. Convention The two figures fit together: the handout's numbers are for teaching the behavior, VCA's for practicing a settle the dog already knows while you sit nearby.
| Stage | What you reward | Where the treat lands | Move on when |
|---|---|---|---|
| Going to the mat | A paw, then two, three, four on the mat | Off the mat (a reset) | The dog steps on with all four paws as soon as the treat is eaten |
| Lying on the mat | A down on the mat, lured or offered | On the mat, between the front paws | The dog lies down on the mat 8 times out of 10 without a lure |
| Adding the word | Going to the mat and lying down after the word | On the mat | The dog goes to the mat on the word from a few feet |
| Staying | Lying there for longer stretches | On the mat, during the down | See the duration ladder below |
| Settling | Signs of relaxation (a hip rolled, a sigh, the chin down) | On the mat, calmly | The dog settles while you do something else nearby |
Duration, distraction and distance
Every stay, wait or settle has three settings. Duration is how long the dog holds the position. Distraction is what else is happening: a person walking past, a ball bouncing, the doorbell. Distance is how far away you are. Trainers call them the three Ds. The AKC's page on the three Ds gives the rule that matters most: work on one D at a time, and when you raise one, drop the others back. Its example: when you work on distance, lower duration back to one second and keep distractions to a minimum. Convention
The reason is simple. Each D is a separate thing the dog has to learn. If you add two seconds and three feet and a passing person all at once, and the dog gets up, you do not know which change was too much. Raise one, and the record tells you exactly what the dog can do.
Which D comes first
The sources agree that duration comes first and that you start with about one second. They disagree about the order of the other two.
- The order we use: duration, then small distractions, then distance. The AKC's stay page builds duration first, adds distractions only after the dog holds a stay for at least 30 seconds, and leaves distance for last. Karen Pryor Clicker Training's stay page says to train duration before distance and to add small distractions, such as waving your arms or bouncing a ball, before distance. Preference
- The other order: duration, then distance, then distraction last. The AKC's three-Ds page says to leave distraction for last and to make sure the dog can handle duration and distance first. The AKC's "go to your place" page follows the same order: duration, then distance, then distractions. Preference
Two pages from the same publisher give different orders, and neither cites a study. We use the first because two independent trainers' pages share it and because small, controlled distractions (you shifting your weight, reaching for a cup) happen in real life before you ever walk away from a dog. If your dog finds your movement harder than your distance, swap them. Your sessions sheet settles it for your dog. Test it
A duration ladder
Build duration in small steps, and reward before the dog gets up, not after. The AKC's stay page starts at one second, then a second or two, and says that once you reach about 10 seconds you can feed several times during the stay. The Today's Veterinary Practice handout says to reward a little before the dog's limit (if the dog can hold 30 seconds, start rewarding at 20) and to raise the time in steps of 5 to 10 seconds. The AKC's three-Ds page gives an example of a 60-second sit with a small treat every 20 seconds. Convention
| Step | Seconds before the release word | Treats during the hold | Notes |
|---|---|---|---|
| 1 | 1 | One, then release | AKC's starting point |
| 2 | 3 | One, then release | The Start here first target |
| 3 | 5 | One or two | |
| 4 | 10 | Two or three, spread out | The Start here Done when; AKC: from about 10 seconds, feed several times during the stay |
| 5 | 15 to 20 | Every 5 seconds or so | Steps of 5 to 10 seconds from here (handout) |
| 6 | 30 | Every 10 seconds or so | AKC's stay page adds distractions at this point |
| 7 | 60 | Every 20 seconds | AKC's three-Ds example |
| 8 | 2 to 5 minutes | Spaced further apart, then a chew on the mat | The handout adds distance once the dog holds about 3 to 5 minutes |
Then, for each new distraction, drop back down the ladder. Then, for distance, drop back again: one step away and back, then two steps, then across the room. The AKC's stay page starts distance with a single step back and a return to treat, then two steps. The Today's Veterinary Practice handout moves the mat a foot further at a time (from 3 feet to 4) and shortens the wait each time it does.
The record
Karen Pryor Clicker Training's stay page gives a two-word instruction that is the point of this manual's log: keep records. With them, the page says, you know how reliable the dog is in a given place. Log each session in the sessions sheet with skill "Mat settle" or "Stay", the place, the distraction level, attempts, successful responses, and the longest hold in seconds in the notes ("longest settle 10 s, standing beside"). The cue inventory carries one row per behavior (mat, stay, leave it, drop it), each with its stage and places. Test it
Count a success as a hold that ended with your release word. A dog that got up first is not a failure of character; it is information that the step was too big.
Stay, wait and the release word
The release word
A release word is the word that tells the dog the job is done and it may move. Without one, the dog has to guess when a stay ends, and most dogs guess "when the handler looks away" or "when the treat stops". The AKC's stay page lists "okay", "free" and "release" as common choices and suggests using the word only to end a behavior. It also warns that "okay" is so common in everyday speech that you may release the dog by accident. Pick a word you do not say in passing, write it on the household agreement in the training log, and use it to end every stay, settle and wait. First behaviors introduced it; this chapter puts it to work. Preference
Say the release word, then invite the dog off the mat with a step back or a tossed piece of food. The release itself is not rewarded with a treat; the AKC's place page says not to click or reward the release. Moving freely is its own reward, and paying for the release teaches the dog to want it sooner.
Stay versus wait
Both words ask the dog to hold still, and households often use them interchangeably. It helps to keep them apart.
| Stay | Wait | |
|---|---|---|
| What it means | Hold this position, here, until I release you | Pause here for a moment; something else is coming |
| How it ends | The release word, usually after you return to the dog | The release word, or the next cue ("let's go", a recall) |
| Typical use | The mat during a meal; a sit for a guest's greeting | At an open door, the car door, the top of the stairs, before the food bowl goes down |
| Position | Sit, down or the mat, as asked | Any position; the dog just does not go through |
| Duration | Built in seconds, then minutes | A few seconds |
Karen Pryor Clicker Training's stay page offers a third idea that suits the mat: a boundary stay, in which the dog learns to stay on a mat, a platform or behind a line rather than to freeze in one position. A dog on a boundary stay may shift from a sit to a down or roll onto its side; what it does not do is leave the mat. For a settle, that is exactly what you want. A relaxed dog moves. Preference
When the dog gets up
It will. The AKC's stay page is clear on the response: do not punish the dog; repeat the cues and start again from an easier step. In practice: say nothing, or a neutral "oops", calmly guide the dog back to the mat with your hand target or a lure, and make the next hold shorter. If the dog breaks twice in a row at the same step, go back a step for the rest of the session.
A dog that gets up repeatedly may be telling you something else. A hard floor can be uncomfortable for an older or bony dog; a mat in a busy corridor is hard to relax on; a dog that needs to go out will not settle. Check those before you blame the training. A dog that cannot settle anywhere, even when nothing is happening, may need the slower approach described under Go deeper, and a dog that is restless and anxious most of the day is a question for the veterinarian first.
Wait at doors and the car
Teach the wait with the door barely open. Ask for a sit or just stand with the dog behind you, open the door an inch, and close it again if the dog moves forward. When the dog stays back while the door is open an inch, mark and treat behind the threshold, then release through the door. Open it a little wider each time. At the car, the same shape: the door opens, the dog holds for a second, the release word lets it jump in or out. A wait at the car door is a safety habit on any street with traffic, so it is worth the practice. Raise one D at a time here too: a wider gap is distance, a longer pause is duration, a person passing outside is distraction.
Stay with distance
Once a stay holds 30 seconds with small distractions, add distance. Step back once, step back in, treat, then release. Two steps. Then sideways, then behind the dog, then out of sight for a second. Each new direction counts as a new step. Return to the dog to feed during the stay; calling the dog to you from a stay mixes the stay up with the recall, and many trainers keep the two apart until both are solid.
Leave it and drop it
Leave it asks a dog not to take something it has noticed: food on the floor, a dropped pill, a chicken bone on the sidewalk. Drop it asks a dog to let go of something it already has in its mouth. Both are taught with rewards, and both depend on the dog trusting that giving something up gets it something better. That trust is the reason this section starts with a stop.
Check for guarding before you start. Watch your dog eat a meal and chew a favorite chew from a little distance, and walk past once. A dog that stiffens, goes still, stares at you, eats faster or gulps as someone approaches its food, a chew or a stolen object, or that growls, snarls, snaps or lunges, is guarding. VCA's page on food bowl aggression lists these subtle signs, and ASPCA's food-guarding page lists stiffening, gulping, freezing, growling and staring. The leave it and drop it exercises in this chapter are not for that dog. ASPCA's page says not to attempt its own exercises if you think the dog is likely to bite, and VCA's says that a dog growling, snarling, snapping or lunging near the dish needs a behavior professional before any exercise. Use management (feed the dog alone behind a closed door or gate, pick up chews and stolen objects only when the dog is elsewhere) and contact your veterinarian, then a veterinary behaviorist or a credentialed behavior consultant. Proof it and keep going says what each credential means. Do not test the dog by reaching for its food; VCA's page says not to let anyone do that.
Resource guarding is a dog's use of body language or aggression to keep people or other animals away from something it values. It is common and understandable from the dog's point of view, and it is outside what this manual teaches. The difference between leave it and guarding is the difference between a dog choosing to look away from food because looking away pays, and a dog defending food because it expects to lose it. You teach the first. A professional works on the second, in person.
Leave it, in stages
The AKC's leave it page teaches the behavior in four steps, then takes it outside. The version below follows its shape. Preference
- Covered by your hand. Place a treat on the floor and cover it with your hand. The dog will sniff, lick and paw. Wait. The moment it stops trying and looks away or backs off, mark, and reward with a different, better treat from your other hand or pocket. The dog does not get the covered treat.
- Uncovered. Once the dog leaves the covered treat reliably, lift your hand a little, ready to cover the treat again if the dog moves toward it. Mark and reward from your other hand when it holds back.
- Standing, with your foot. With the dog on a leash, stand up and drop the treat, ready to cover it with your foot. The AKC page uses the leash to keep the dog from reaching the food.
- Adding the word. Say "leave it" before you drop the food, then mark and reward when the dog leaves it alone.
Outside, the AKC page lays low-value treats in a row several feet apart while the dog is out of the room, then walks the dog past them on a leash, rewarding at each one, and finally walks past the whole row after a single "leave it", with something special (a game of tug, a chew) for success.
Two rules from the same page keep the cue working. First, prevent the dog from getting the food during practice; if it grabs a treat even once, it has learned that leave it sometimes ends with the treat. Cover it with your foot, use the leash to block, and set up the next repetition to be easier. Second, do not use the cue where you know the dog will not respond; go back to training with that kind of distraction instead. Every failed "leave it" makes the next one weaker. Preference
The reward for leaving something comes from you, not from the thing left. That is a deliberate choice. A dog that is sometimes allowed the dropped food after a "leave it" learns to wait for permission and then dive; a dog that is paid from your hand every time learns to turn toward you. Some trainers pair leave it with a separate "take it" cue for things the dog may have, and VCA's stealing page mentions teaching both.
Drop it, as a trade
Drop it is taught as a trade: the dog gives up what it has and gets something better, and often gets the item back. The AKC's drop it page and Humane World for Animals' chewing page describe the same shape. Preference
- Start with a toy the dog likes a little. The AKC page suggests a low-value toy, interesting but not exciting, so giving it up is easy. Let the dog play with it for a few seconds.
- Offer the trade. Hold a high-value treat at the dog's nose. When the toy drops, mark and give the treat.
- Pick up the toy while the dog eats. Then give the toy back and repeat. Getting the toy back is part of what makes dropping it worth it.
- Add the word. When the dog drops the toy as soon as the treat appears, say "drop it" (or "trade", "out", "give") just before you show the treat.
- Show the treat later. After several repetitions, say the cue and wait one second before showing the treat, then a little longer. When the dog drops on the word alone, give a big reward.
Humane World for Animals' chewing page gives the same advice in everyday terms: if the dog has something it should not, trade it for a toy or a treat, and add a "give" cue as the dog catches on. Its page says that removing items from a dog's mouth can lead to guarding. The AKC's page says not to pry the dog's mouth open, because it risks a bite and can make the dog swallow the item, and not to chase a dog that runs off with something, because chasing raises the chance it swallows the object to keep it. Humane World for Animals' page adds that chasing gives the dog what it wants. VCA's stealing page says to avoid the chase game and not to reach for the dog in anger after it takes something. Convention
If the dog runs off with the toy during practice, the AKC page suggests practicing on a leash. If it tries to grab the toy back before you do, toss a treat a few feet away to buy time. The AKC page notes that a reliable drop it can be what gets a dropped pill or a piece of chocolate out of a dog's mouth, and that the dog should not be given dangerous items back. If your dog swallows something that worries you, call your veterinarian.
An adult should run drop it practice. It involves reaching toward items close to the dog's mouth, and ASPCA's food-guarding page says its guarding exercises are not for anyone under 18. Children can help with leave it by dropping food when an adult says so, with the dog on a leash held by the adult.
| Leave it | Drop it | Guarding (not taught here) | |
|---|---|---|---|
| The dog has the item | No, it has noticed it | Yes, in its mouth | Yes, or near it |
| What the dog does | Looks away, turns toward you | Lets go for a trade | Stiffens, stares, eats faster, growls, snaps |
| What you do | Cover, wait, mark, pay from your hand | Offer the trade, mark, pay, often give the item back | Manage, stop, call the veterinarian and a behavior professional |
| Who practices | An adult; a child can drop food under supervision | An adult | A professional, in person |
Drop it with toys, and the trade game
Once drop it works with a low-value toy, raise the value one step at a time: a better toy, then a tug toy during a game, then a chew the dog likes. Play tug, say "drop it", trade for a treat, then restart the game as the reward. For many dogs the restarted game is worth more than the treat, and the cue becomes part of play. Keep the item the dog gives up within reach and give it back most of the time, so dropping it is rarely a loss. For something you cannot give back (a sock, a dropped pill), trade for something especially good, and practice with harmless items far more often than you ever need the cue for real.
Leave it on walks
On walks, the leave it you practiced at home meets a sidewalk full of smells and scraps. Keep the leash short near known hazards (the alley behind a restaurant, a picnic area) and say "leave it" a step or two before the dog reaches the item, while it can still choose. If the dog's nose is already on the scrap, the cue is too late; use the leash to move on and set up an easier version at home.
Doors, bowls and impulse control
Impulse control is the everyday name for a dog holding back from something it wants until it is invited: waiting at a door, not snatching a treat from a hand, leaving the food bowl alone until it is down. It is not a single trait the dog has or lacks. It is a set of specific behaviors, each taught in a specific situation with rewards, and the three Ds apply to each. A dog that waits beautifully at the back door may bolt through the front door, because nobody practiced the front door.
Not "nothing in life is free"
One popular program, often called "nothing in life is free", asks the dog to perform a cue (usually a sit) before it gets anything it wants: meals, walks, attention, a door opening. Some versions are framed as making the dog earn its place or accept the owner's leadership. This manual does not use that framing, for two reasons.
First, the leadership idea comes from the dominance model that How dogs learn explains and this manual does not use. Second, the food-related claims do not have evidence behind them. VCA's 2025 page on food bowl aggression, written by veterinary behaviorists, says there is no scientific evidence that hand-feeding a dog, or asking it to follow a long series of cues before its meal, will help treat or prevent food-related aggression. The same page warns that a dog waiting for each piece of food to be handed over may become more frustrated. Convention
What this manual does instead is narrower and more practical. A wait before the bowl goes down is a useful household manner, taught like any other wait: one second, then two, then the release word, and the bowl goes down. It is not a test, and it is not a guarding fix. VCA's page says not to remove a dog's bowl, or the food in it, while the dog is eating. A wait at the door is a safety behavior taught at the door. A sit before a guest greets the dog is a greeting behavior taught with the guest. Each one is taught on its own, rewarded, and recorded.
| Situation | Behavior taught | Taught with | What it is not |
|---|---|---|---|
| Front door | Wait until released | Door opened an inch at a time, treat behind the threshold | A test of who goes first |
| Food bowl | Wait a few seconds while the bowl goes down | Release word, then the meal | A guarding fix; a reason to take food away |
| Guests | Go to the mat, or sit for a greeting | Mat settle practiced before guests come; a leash or gate while learning | Proof of a calm temperament |
| Meals at the table | Settle on the mat nearby | A chew or food toy on the mat, counted against the allowance | Punishment for begging |
| Kitchen counters | Four feet on the floor; leave it | Management first (nothing on counters, a gate), then rewards for the floor | Something to correct after the fact |
VCA's predictability page adds one more piece: a chew or food toy given on the mat makes the mat the best place in the room, and any food in it counts against the day's calories. Count it in the treat budget.
The relaxation protocol, benchmarks and the 80-percent rule Go deeper
This section is optional on a first pass. It names a longer method for dogs that find it hard to settle, gives a show benchmark for sit and down stays, and gives one practitioner's rule for when to raise the difficulty.
The relaxation protocol
Karen Overall, a veterinary behaviorist, published a Protocol for Relaxation in her 2013 textbook on clinical behavioral medicine, and her site hosts the chapter. It is under copyright, so we describe only its shape. It is a series of 15 task sheets that ask the dog to sit and stay while the handler does gradually harder things nearby (moving, stepping away, making sounds), and it rewards the dog for looking relaxed, not merely for sitting. The protocol says a dog should be able to sit for about 15 seconds first, describes working through the task sheets over at least two weeks at the fastest pace, and suggests sessions of about 15 to 20 minutes once or twice a day while also saying any schedule the owner can keep is better than one they resent. It is one practitioner's method, built on a sit rather than a mat. Preference
The 2025 Today's Veterinary Practice client handout links the Overall protocol and gives shorter exercises with session numbers: a mat exercise, a "relax" or "settle" exercise that rewards signs of relaxation in a down, and a "breathe" exercise that marks a deep inhale. Its relax exercise builds the relaxed down from 2 seconds to 3, 5, 10, 15 and 30, and goes back at least 5 seconds whenever the dog cannot hold the step. Its sessions are 2 to 5 minutes, 2 to 3 times a day. If your dog's mat settle stalls at a few seconds because the dog cannot relax, the handout is the next thing to read, and the Overall protocol is the longer version. A dog that cannot settle anywhere, or that is anxious much of the day, is a question for your veterinarian before either. Convention
Show benchmarks: 1-minute sit, 3-minute down
The University of Arizona Cooperative Extension publishes obedience guidelines for the Pima County 4-H dog project (revised December 2023). Its first-year class includes a long sit of 1 minute and a long down of 3 minutes, with the handlers at the end of a six-foot leash, facing their dogs, and all the dogs in the class doing it together. These are show rules, and the same guidelines require equipment and handling this manual does not use; we take only the times, as a benchmark you can compare your record with. A dog that holds a 3-minute down on its mat with you six feet away, and other people moving about, has a stay worth writing down. Convention
The AKC's Canine Good Citizen test, item 6, asks for a sit and a down on cue and a stay while the handler walks to the end of a 20-foot line and back. Come when called covers item 7, the recall that follows on the same line.
The 80-percent rule
Karen Pryor Clicker Training's page on proofing, by a certified trainer, gives a rule for when to make an exercise harder: raise the criteria only when the dog succeeds about 80 percent of the time at the current level, and when the dog succeeds less than 50 percent of the time, break the exercise into smaller steps. The page also suggests working on one aspect at a time in a session, and says most dogs would prefer five three-minute sessions a day to one session of ten or fifteen minutes. Preference
The sessions sheet records attempts and successful responses, and its percent (successful responses divided by attempts) makes the rule easy to apply: 80 percent or more, raise one D; 50 to 79 percent, stay where you are; under 50 percent, go back a step. Ten attempts make the arithmetic easy: 8 of 10 is 80 percent. The cue inventory's own standard (five out of five before a stage counts) is stricter, on purpose: the inventory records what the dog does reliably, while the 80-percent rule decides what to practice next. Proof it and keep going uses both when it builds the proofing plan. Test it
See it: Pearl and the condo kitchen See it
Pearl, leave it and a mat by the chair. Walter Kaminski is retired and lives alone. In August 2026 he and Pearl moved to a condominium from the house he had lived in for 12 years. Pearl is a 9-year-old beagle, 28 lb, with Walter since she was a puppy. She had no formal training and knows "sit" for a biscuit. Her hearing is fading. Walter's goals are a calm settle, leave it (she steals food from the counter at the new place) and comfort when alone. He chooses 10 minutes twice a day.
Known. Walter's test in First behaviors showed that Pearl responds to hand signals at any distance and to words only up close, so the cue inventory records hand signals as primary. Pearl eats a senior diet from her veterinarian, whose figure is 520 kcal a day, so her extras budget is 52 kcal a day. Her kibble is 340 kcal a cup, and any kibble used in sessions is taken out of her meals, not counted against that budget. She has not bitten.
Assumed. Walter watched Pearl eat her dinner from across the room and walked past her twice while she chewed a biscuit, and saw none of the guarding signs in the Limits box above. On that observation, which is Walter's and is provisional, leave it and drop it are for her. Had he seen her stiffen or eat faster, the exercises would have stopped there.
The plan. Management first: nothing edible stays on the counter, and a gate closes the kitchen when Walter cannot watch, so the counter stops paying while she learns. For the mat, he chooses a thick, soft bed beside his armchair, because a 9-year-old beagle may find a thin mat on a hard condo floor uncomfortable. He marks with a thumbs-up she can see, lures the down, and feeds between her paws, using her kibble allowance. His release signal is a sweep of the arm toward the room, with the word "free" said up close. Leave it starts with a treat under his hand on the floor, and his signal for it is a flat palm turned toward her.
The first week's record (fictional). Day 1: mat, 10 paws-on repetitions, then 3 seconds lying down. Day 3: 10 seconds standing beside the bed, 9 of 10. Day 5: leave it with the treat covered, 8 of 10; uncovered, 6 of 10, so he stays at uncovered. Day 7: 30 seconds on the bed while he sits in the chair and turns a page. The cue inventory shows "mat: on signal, living room" and "leave it: on signal, kitchen floor, covered and uncovered".
Open, and who settles it. Whether Pearl's new barking when left has a medical cause or is a response to the move is not this chapter's question; her veterinarian found nothing, suggested the alone-time ladder from Crate, pen and alone time and a recheck in a month, so it stays provisional. A mat settle may help the evenings, but it is not a treatment for distress when alone, and Walter does not count it as one. Whether Pearl will leave food on the counter when Walter is out of the room is open; the gate stays until the log shows leave it holding in the kitchen with him at a distance.
- Karen Pryor Clicker Training: Teach Settle on a Mat: the shaping route to a mat, from one paw to a down, with no lure.
- Today's Veterinary Practice: Teaching Relaxation Exercises (PDF handout): a veterinary client handout with the mat, settle and breathe exercises and session lengths.
- AKC: "Leave It": the four indoor steps and the walk past a row of treats, as used in this chapter.
- VCA: Food Bowl Aggression in Dogs: the subtle guarding signs and why the exercises wait for a professional when they appear.
Practice Practice
Minimum: mat on, down on the mat, ten seconds beside it ending with the release word, and leave it with a covered treat if the dog shows no guarding signs; the longest settle written in the log. Time: about four sessions of 2 to 5 minutes over a few days, then a few minutes a day. Materials: a mat, towel or dog bed that can move; small treats counted against the allowance; a leash for leave it on your feet; the training log.
- Check for guarding. Watch the dog eat and chew from a distance and walk past once. If you see any of the signs in the Limits box, skip leave it and drop it, keep the mat work, and use the professional route. Note what you saw.
- Get the dog onto the mat. Mark and treat each paw on the mat, tossing the treat off so the dog comes back. About ten times. Then four paws only.
- Lure the down on the mat. When the dog stands on the mat, lure or cue the down and feed between its front paws. Stop luring once the dog lies down on the mat 8 times out of 10.
- Build duration. Stand beside the mat. Feed during the down at 3 seconds, then 5, then 10, and end each hold with the release word. If the dog gets up, guide it back and make the next one shorter.
- Start leave it. Cover a treat on the floor with your hand. Mark the moment the dog looks away or backs off; pay from your other hand. Ten repetitions, then lift your hand a little.
- Record. Log the session in the sessions sheet (skill "Mat settle" or "Leave it", place, attempts, successful responses, longest hold in the notes) and update the cue inventory rows for mat and leave it.
After the first ten seconds. Climb the duration ladder to 30 seconds with you standing beside the mat. Then add one small distraction at a time (shift your weight, reach for a cup, take one step sideways), dropping back to 5 seconds each time. Then add distance, one step at a time, dropping duration and distraction back. Add the cue word for the mat once the dog goes there on its own as you approach it.
Drop it. Start the trade with a low-value toy, as in the drop it steps above. Ten trades, giving the toy back each time. Add the word when the toy drops as soon as the treat appears.
A settle during a real event. When the mat holds a minute with you nearby, use it during a meal or a phone call, with a chew on the mat. Before guests come, practice with one calm visitor and the dog on a leash or behind a gate while it learns.
Your dog Your dog
Check your work Check your work
Criteria
- Did you check for guarding signs before leave it and drop it, and write down what you saw?
- Does every treat for staying land on the mat, during the down?
- Does every hold end with the release word, rather than the dog deciding?
- Does the record show one D raised at a time, with the others dropped back?
- Is the longest settle written in seconds, with where you were standing?
- Does leave it pay from your hand, with the dog kept from the left item during practice?
- Is drop it taught as a trade, with the item given back during practice and no chasing or prying?
Worked example
Fictional example. Pearl's record shows the guarding check, with Walter's note that he saw none of the signs. Every mat treat lands between her paws. Every hold ends with the arm sweep and "free". The log raises duration first (3, then 10, then 30 seconds) and has not yet added distance or distraction beyond Walter turning a page. The cue inventory shows the mat and leave it at "on signal" in one room each, which is accurate: neither has been practiced anywhere else. Against the criteria, Pearl's record meets the Done when (10 seconds beside the bed) and is ready for the first distraction step. Drop it has not started, and the counter gate stays.
Common mistakes
- Feeding off the mat during the stay. If the dog gets up just before each treat, check where the treat lands.
- Waiting for the dog to break. If most holds end with the dog getting up, the reward is coming too late; reward a little before the dog's limit.
- Raising two Ds at once. If the dog was solid at 30 seconds and fails when you step back, check whether you kept the 30 seconds while adding the step.
- "Okay" as the release word. If the dog leaves the mat while you talk on the phone, listen for the word.
- Letting the dog have the left item. If leave it is getting slower, check whether the dog has been allowed the item after waiting.
- Grabbing the item for drop it. If the dog has started running off with things, check whether anyone has been chasing it or taking things from its mouth.
- Missing guarding signs. Eating faster and going still as you pass are the quiet ones. If you are not sure what you saw, treat it as a sign and ask.
What is still unresolved
This chapter cannot tell you whether your dog guards things; your observation, and a professional if you see signs, settle that. It cannot tell you whether a dog that cannot settle anywhere has a medical or anxiety problem; the veterinarian settles that first. Whether a settle or a leave it holds in a new room, with guests or on a walk is settled by practicing there and writing it down. Everyday problems uses leave it and the mat in problem plans for jumping and stealing, and Proof it and keep going proofs them across places.